A Novel Approach Based on Marine Predators Algorithm for Medical Image Enhancement

One of the approaches to improve the performance of image processing applications is to enhance the input image quality. Some common problems with medical images include lack of sharpness, noise, or low contrast. In this study, we propose a new algorithm to improve the quality of brain Magnetic resonance images (MRI). Firstly, we use image enhancement algorithms such as Contrast limited adaptive histogram equalization (CLAHE), Denoise based on convolutional neural network (DN-CNN), and Laplacian edge detection (LED) to create I_1 , I_2 , and I_3 images from the input image ( I ), respectively. Secondly, the Marine predators algorithm (MPA) is used to find the adaptive parameters β _1 , β _2 , and β _3 corresponding to I_1 , I_2 , and I_3 . Finally, the enhanced image is made up of the sum of the I_1 , I_2 , and I_3 images multiplied by the coefficients β _1 , β _2 , and β _3 , respectively. Experiments show that the images produced by our model are better than those produced by some recent image enhancement methods. Furthermore, our model also allows for improving the performance of image fusion algorithms.